* RFA: Please comment on this ia64 patch
@ 2003-05-02 15:30 H. J. Lu
2003-05-02 23:43 ` Jim Wilson
0 siblings, 1 reply; 3+ messages in thread
From: H. J. Lu @ 2003-05-02 15:30 UTC (permalink / raw)
To: binutils
I added the support for 'o' and "unwind" in section directive for ia64
in this patch:
http://sources.redhat.com/ml/binutils/2003-04/msg00464.html
Could someone please comment on it?
Yes, "unwind" is missing from the Intel assembler manual.
Thanks.
H.J.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: RFA: Please comment on this ia64 patch
2003-05-02 15:30 RFA: Please comment on this ia64 patch H. J. Lu
@ 2003-05-02 23:43 ` Jim Wilson
2003-05-03 16:05 ` H. J. Lu
0 siblings, 1 reply; 3+ messages in thread
From: Jim Wilson @ 2003-05-02 23:43 UTC (permalink / raw)
To: H. J. Lu; +Cc: binutils
H. J. Lu wrote:
> I added the support for 'o' and "unwind" in section directive for ia64
> in this patch
The part I can't answer is the question about having a generic letter
that maps to SEC_LINK_ORDER. Someone else would have to comment on
that. I suspect it wouldn't be very useful at this time because
GNU/Linux doesn't actively use SEC_LINK_ORDER.
I noticed you didn't put 'o' into the IA-64 error message. If we are
accepting 'o', we should document it in the error message.
I noticed you don't set the SHF_LINK_ORDER when seeing the 'o'. This
does do something useful, though maybe not for GNU/Linux system. I know
that the IA-64 ABI and HPUX require that it be set. This has come up
before. Note that bfd/elfxx-ia64.c tries to set SHF_LINK_ORDER and
SHT_IA_64_UNWIND for unwind section. So I think your patch should
return SHF_LINK_ORDER for 'o' to be consistent. See also the code in
elfNN_ia64_final_write_processing that sets the sh_link and sh_info
fields which is supposed to make SHF_LINK_ORDER useful.
Otherwise the patch looks fine. The non-IA-64 parts look like obvious
bug fixes, so I'll approve the entire patch with the two suggestions I
made above.
Jim
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: RFA: Please comment on this ia64 patch
2003-05-02 23:43 ` Jim Wilson
@ 2003-05-03 16:05 ` H. J. Lu
0 siblings, 0 replies; 3+ messages in thread
From: H. J. Lu @ 2003-05-03 16:05 UTC (permalink / raw)
To: Jim Wilson; +Cc: binutils
[-- Attachment #1: Type: text/plain, Size: 1356 bytes --]
On Fri, May 02, 2003 at 04:44:14PM -0400, Jim Wilson wrote:
> H. J. Lu wrote:
> > I added the support for 'o' and "unwind" in section directive for ia64
> > in this patch
>
> The part I can't answer is the question about having a generic letter
> that maps to SEC_LINK_ORDER. Someone else would have to comment on
> that. I suspect it wouldn't be very useful at this time because
> GNU/Linux doesn't actively use SEC_LINK_ORDER.
>
> I noticed you didn't put 'o' into the IA-64 error message. If we are
> accepting 'o', we should document it in the error message.
>
> I noticed you don't set the SHF_LINK_ORDER when seeing the 'o'. This
> does do something useful, though maybe not for GNU/Linux system. I know
> that the IA-64 ABI and HPUX require that it be set. This has come up
> before. Note that bfd/elfxx-ia64.c tries to set SHF_LINK_ORDER and
> SHT_IA_64_UNWIND for unwind section. So I think your patch should
> return SHF_LINK_ORDER for 'o' to be consistent. See also the code in
> elfNN_ia64_final_write_processing that sets the sh_link and sh_info
> fields which is supposed to make SHF_LINK_ORDER useful.
>
> Otherwise the patch looks fine. The non-IA-64 parts look like obvious
> bug fixes, so I'll approve the entire patch with the two suggestions I
> made above.
>
Thanks. This is the patch I checked in.
H.J.
[-- Attachment #2: gas-ia64-unwind.patch --]
[-- Type: text/plain, Size: 5015 bytes --]
gas/
2003-05-03 H.J. Lu <hjl@gnu.org>
* config/obj-elf.c (obj_elf_parse_section_letters): Make it a
fatal error for unknown section attribute.
* config/tc-alpha.c (alpha_elf_section_letter): Return -1 for
unknown section attribute.
* config/tc-ia64.c (ia64_elf_section_letter): Likewise.
* config/tc-ppc.c (ppc_section_letter): Likewise.
* config/tc-ia64.c (ia64_elf_section_letter): Handle 'o'.
(ia64_elf_section_type): Accept "unwind".
gas/testsuite/
2003-05-03 H.J. Lu <hjl@gnu.org>
* gas/ia64/ia64.exp: Add unwind.
* gas/ia64/unwind.s: New. Test the new section attribute 'o'
and the new section type "unwind".
* gas/ia64/unwind.d: Likewise.
--- gas/config/obj-elf.c.unwind 2003-05-03 08:51:47.000000000 -0700
+++ gas/config/obj-elf.c 2003-05-03 08:51:54.000000000 -0700
@@ -845,10 +845,7 @@ obj_elf_parse_section_letters (str, len)
attr |= md_attr;
else
#endif
- {
- as_warn ("%s", bad_msg);
- attr = -1;
- }
+ as_fatal ("%s", bad_msg);
}
break;
}
--- gas/config/tc-alpha.c.unwind 2003-05-03 08:51:47.000000000 -0700
+++ gas/config/tc-alpha.c 2003-05-03 08:51:54.000000000 -0700
@@ -5635,7 +5635,7 @@ alpha_elf_section_letter (letter, ptr_ms
return SHF_ALPHA_GPREL;
*ptr_msg = _("Bad .section directive: want a,s,w,x,M,S,G,T in string");
- return 0;
+ return -1;
}
/* Map SHF_ALPHA_GPREL to SEC_SMALL_DATA. */
--- gas/config/tc-ia64.c.unwind 2003-05-03 08:51:47.000000000 -0700
+++ gas/config/tc-ia64.c 2003-05-03 08:56:20.000000000 -0700
@@ -968,9 +968,11 @@ ia64_elf_section_letter (letter, ptr_msg
{
if (letter == 's')
return SHF_IA_64_SHORT;
+ else if (letter == 'o')
+ return SHF_LINK_ORDER;
- *ptr_msg = _("Bad .section directive: want a,s,w,x,M,S,G,T in string");
- return 0;
+ *ptr_msg = _("Bad .section directive: want a,o,s,w,x,M,S,G,T in string");
+ return -1;
}
/* Map SHF_IA_64_SHORT to SEC_SMALL_DATA. */
@@ -1004,6 +1006,9 @@ ia64_elf_section_type (str, len)
if (STREQ (ELF_STRING_ia64_unwind_once))
return SHT_IA_64_UNWIND;
+ if (STREQ ("unwind"))
+ return SHT_IA_64_UNWIND;
+
if (STREQ ("init_array"))
return SHT_INIT_ARRAY;
--- gas/config/tc-ppc.c.unwind 2003-05-03 08:51:47.000000000 -0700
+++ gas/config/tc-ppc.c 2003-05-03 08:51:54.000000000 -0700
@@ -2754,7 +2754,7 @@ ppc_section_letter (letter, ptr_msg)
return SHF_EXCLUDE;
*ptr_msg = _("Bad .section directive: want a,e,w,x,M,S,G,T in string");
- return 0;
+ return -1;
}
int
--- gas/testsuite/gas/ia64/ia64.exp.unwind 2003-05-03 08:51:47.000000000 -0700
+++ gas/testsuite/gas/ia64/ia64.exp 2003-05-03 08:52:29.000000000 -0700
@@ -47,4 +47,5 @@ if [istarget "ia64-*"] then {
run_dump_test "order"
run_dump_test "global"
run_dump_test "secname"
+ run_dump_test "unwind"
}
--- gas/testsuite/gas/ia64/unwind.d.unwind 2003-05-03 08:51:54.000000000 -0700
+++ gas/testsuite/gas/ia64/unwind.d 2003-05-03 08:51:54.000000000 -0700
@@ -0,0 +1,30 @@
+#readelf: -S
+#name: ia64 unwind section
+
+There are 9 section headers, starting at offset 0xa0:
+
+Section Headers:
+ \[Nr\] Name Type Address Offset
+ Size EntSize Flags Link Info Align
+ \[ 0\] NULL 0000000000000000 00000000
+ 0000000000000000 0000000000000000 0 0 0
+ \[ 1\] \.text PROGBITS 0000000000000000 00000040
+ 0000000000000000 0000000000000000 AX 0 0 16
+ \[ 2\] \.data PROGBITS 0000000000000000 00000040
+ 0000000000000000 0000000000000000 WA 0 0 1
+ \[ 3\] \.bss NOBITS 0000000000000000 00000040
+ 0000000000000000 0000000000000000 WA 0 0 1
+ \[ 4\] \.IA_64\.unwind_inf PROGBITS 0000000000000000 00000040
+ 0000000000000008 0000000000000000 A 0 0 8
+ \[ 5\] \.IA_64\.unwind IA_64_UNWIND 0000000000000000 00000048
+ 0000000000000008 0000000000000000 AL 1 1 8
+ \[ 6\] \.shstrtab STRTAB 0000000000000000 00000050
+ 000000000000004d 0000000000000000 0 0 1
+ \[ 7\] \.symtab SYMTAB 0000000000000000 000002e0
+ 0000000000000090 0000000000000018 8 6 8
+ \[ 8\] \.strtab STRTAB 0000000000000000 00000370
+ 0000000000000001 0000000000000000 0 0 1
+Key to Flags:
+ W \(write\), A \(alloc\), X \(execute\), M \(merge\), S \(strings\)
+ I \(info\), L \(link order\), G \(group\), x \(unknown\)
+ O \(extra OS processing required\) o \(OS specific\), p \(processor specific\)
--- gas/testsuite/gas/ia64/unwind.s.unwind 2003-05-03 08:51:54.000000000 -0700
+++ gas/testsuite/gas/ia64/unwind.s 2003-05-03 08:51:54.000000000 -0700
@@ -0,0 +1,4 @@
+ .section .IA_64.unwind_info, "a", "progbits"
+ data8 1234
+ .section .IA_64.unwind, "ao", "unwind"
+ data8 1234
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-05-03 16:05 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-05-02 15:30 RFA: Please comment on this ia64 patch H. J. Lu
2003-05-02 23:43 ` Jim Wilson
2003-05-03 16:05 ` H. J. Lu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).